INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 0-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.hook_resid_pre

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

torch.float32

Hook Point

blocks.0.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

Reviewed

-0.68

rouse

-0.66

rim

-0.65

APS

-0.64

isse

-0.64

hare

-0.63

cit

-0.63

asp

-0.63

alt

-0.61

 CTRL

-0.61

POSITIVE LOGITS

 Tribal

0.69

Tex

0.63

_-

0.63

bernatorial

0.62

oration

0.60

 Medina

0.59

 Amendments

0.59

 Millenn

0.59

esville

0.59

 Masquerade

0.58

Activations Density 0.000%

No Known Activations

This feature has no known activations.